© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 50814
    Prev
    Next
    INDEX
    Explanations

    the word "especially" and its variants, indicating emphasis or significance in context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    introducing specific reasons or conditions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    getMenuInflater
    -0.41
    WireFormatLite
    -0.39
     acto
    -0.35
    }{*}{}
    -0.35
    UserScript
    -0.35
    INSEE
    -0.34
    AutoModerator
    -0.33
    KURZBESCHREIBUNG
    -0.33
     robot
    -0.33
     actos
    -0.33
    POSITIVE LOGITS
    especially
    0.93
    Especially
    0.91
     especially
    0.91
     specialmente
    0.81
     Especially
    0.80
    特别是
    0.80
    particularly
    0.79
    尤其是
    0.76
     particularly
    0.74
     zwłaszcza
    0.73
    Activations Density 0.013%

    No Known Activations